110 C
.J. Jeffery, A.J. Underwood J. Exp. Mar. Biol. Ecol. 252 2000 109 –127
1. Introduction
Many demographic models in ecology are based on closed populations Roughgarden and Iwasa, 1986 but many marine species such as barnacles have open populations with
sessile adults and long-lived dispersive larval stages. Early models to explain settlement of sessile marine invertebrates, especially barnacles, were based on populations which
had large numbers of larvae arriving to settle Southward and Crisp, 1954, 1956; Barnes, 1956; Connell, 1961a,b; Paine, 1974; Kendall et al., 1985. Because larval supply was
considered to be unlimited, these early models emphasised that competition, predation and physical disturbance in the adult stage determined the abundances of intertidal
organisms Connell, 1961a,b, 1970; Dayton, 1971; Paine, 1974, 1984; Menge, 1976; Menge and Sutherland, 1976; Lubchenco and Menge, 1978; Menge and Lubchenco,
1981. More recent studies have linked barnacle populations with larval supply Grosberg, 1982; Hawkins and Hartnoll, 1982; Underwood and Denley, 1984; Gaines
and Roughgarden, 1985; Kendall et al., 1985.
Where larval settlement is small, however, very different patterns emerge than from areas where larval settlement is large Underwood et al., 1983; Underwood and Denley,
1984, so patterns of distribution could also be due to processes of settlement or recruitment. Other studies support the relationship between larval settlement and
assemblages of species Denley and Underwood, 1979; Grosberg, 1982; Keough, 1983; Caffey, 1985; Connell, 1985; Gaines and Roughgarden, 1985, 1987; Roughgarden et al.,
1985; Roughgarden and Iwasa, 1986; Raimondi, 1988a,b, 1990, 1991; Minchinton and Scheibling, 1991; Bertness et al., 1992; Hurlbut, 1992; Gaines and Bertness, 1993;
Pineda, 1994; Carroll, 1996; Robles, 1997.
Larval delivery or supply Raimondi, 1991 may therefore influence the abundance and distribution of these open populations, so processes influencing the numbers of
larvae arriving in different places will be important. For example, Roughgarden et al. 1991, 1994 have demonstrated that recruitment pulses of barnacles on the Californian
coast are associated with the accumulation of larvae in fronts which are driven shorewards and deposited on the coast when winds relax. Other studies at large spatial
scales have shown a positive relationship with onshore winds Hawkins and Hartnoll, 1982; Bertness et al., 1996 or no relationship Shanks, 1986; Wethey, 1986.
Hydrodynamics at small spatial scales, such as direction of flow, turbulence and shear stress, have also been shown to have roles in determining arrival and settlement of
cyprids Mullineaux and Butman, 1991; Mullineaux and Garland, 1993.
In Botany Bay New South Wales, Australia, there are great spatial variations in densities of the honeycomb barnacle Chamaesipho tasmanica. In 1989, early in this
study, larger numbers of juveniles and adults were recorded on lower than on upper mid-littoral shores. If larval supply is responsible for these variations in abundances it
can be predicted that numbers of larvae settling on the shore should be directly related to the numbers arriving in the plankton. Moreover, if traps were set to catch cyprids at
different levels on the shore at one site, there would be more larvae caught in traps low on the shore than further up the shore. Barnacle cyprids were also observed to arrive in
major pulses in September and October, 1989 close to full moons when strong southerly winds and waves more than 1 m prevailed. If such meteorological patterns are closely
C .J. Jeffery, A.J. Underwood J. Exp. Mar. Biol. Ecol. 252 2000 109 –127
111
associated with pulses of arrival of barnacles, these conditions should also coincide with future peaks of larval supply.
Further, the densities of Chamaesipho differ from those of another species Tetra- clitella purpurascens Wood. Again, if differences in larval supply are responsible,
there should be consistent differences in numbers of larvae of the two species arriving, and these numbers should be consistent with the subsequent numbers of barnacles of the
two species.
More Chamaesipho inhabit lower mid-littoral shores whereas Tetraclitella are evenly distributed across the mid-littoral shore in more shaded areas Denley and Underwood,
1979. These different spatial distributions suggest that different processes are operating. For example, some studies have demonstrated that more larvae may arrive on lower
areas of a shore because they settle here first and settle later on upper areas Roughgarden et al., 1988. Different positions of larvae in the water column Grosberg,
1982; Gaines et al., 1985 may also explain different spatial patterns. Variations in amount of water flowing over different areas may also be responsible for spatial
distributions of these barnacles so that there should be close correlations between patterns of water-flow and numbers of cyprids arriving.
From the initial field observations of differences in distributions of the barnacles, it was predicted that physical processes at large spatial scales lunar cycles, wind direction,
wind speed and wave height and at small spatial scales water-flow would affect larval supply, hence spatial distributions of juveniles on the substratum. These hypotheses were
tested on a shore in New South Wales.
2. Materials and methods
